Causes of faecal and urinary incontinence after total mesorectal excision for rectal cancer based on cadaveric surgery CHAPTER 8

Transcription

1 CHAPTER 8 Causes of faecal and urinary incontinence after total mesorectal excision for rectal cancer based on cadaveric surgery Wallner C, Lange MM, Bonsing BA, Maas CP, Wallace CN, Dabhoiwala, Rutten HJ, Lamers WH, DeRuiter MC, van de Velde CJH. (Both C.W. and M.M.L. contributed equally to this work.) Journal of Clinical Oncology Sep 20;26(27):

2 Chapter 8 ABSTRACT Background Total mesorectal excision (TME) for rectal cancer may result in anorectal and urogenital dysfunction. We aimed to study possible nerve disruption during TME and its consequences for functional outcome. Because the levator ani muscle plays an important role in both urinary and faecal continence, an explanation could be peroperative damage of the nerve supply to the levator ani muscle. Methods Total mesorectal excision was performed on cadaver pelves. Subsequently, the anatomy of the pelvic floor innervations and its relation to the pelvic autonomic innervation and the mesorectum were studied. Additionally, data from the Dutch TME trial were analysed to relate anorectal and urinary dysfunction to possible nerve damage during TME procedure. Results Cadaver TME surgery demonstrated that, especially in low tumours, the pelvic floor innervation can be damaged. Furthermore, the origin of the levator ani nerve was located in close proximity of the origin of the pelvic splanchnic nerves. Analysis of the TME trial data showed that newly developed urinary and faecal incontinence was present in 33.7 percent and 38.8 percent of patients, respectively. Both types of incontinence were significantly associated with each other (p=0.027). Low anastomosis was significantly associated with urinary incontinence (p=0.049). One third of the patients with newly developed urinary and faecal incontinence also reported difficulty in bladder emptying, for which excessive peroperative blood loss was a significant risk factor. Conclusion Peroperative damage to the pelvic floor innervation could contribute to faecal and urinary incontinence after TME, especially in case of a low anastomosis or damage to the pelvic splanchnic nerves. 120

3 The role of levator ani nerve damage in the development of incontinence problems INTRODUCTION The past two decades have witnessed substantial improvement in survival from rectal cancer, resulting from earlier diagnosis, improved efficiency and delivery of radiotherapy, and advances in surgical techniques, such as total mesorectal excision (TME). The TME procedure removes the rectum with its primary lymphovascular field of drainage as an intact package. Under direct vision along pre-existing embryologically determined planes, sharp dissection divides the mesorectal fascia (ie, the visceral fascia surrounding the mesorectum) from the pelvic parietal fascia overlying the pelvic floor, thereby preserving the autonomic nerves required for maintenance of urogenital function. 1-3 However, despite this, clinical studies report a high incidence of pelvic organ dysfunction, and the good functional results achieved by expert rectal surgeons have not yet been reproduced in larger studies. 3,4 Surgical damage to the pelvic autonomic nerves is believed to be an important cause of urinary dysfunction. 5-7 The pelvic parasympathetic supply (pelvic splanchnic nerves or nervi erigentes) arises from sacral nerves S2 to S4, whereas the sympathetic supply is by the hypogastric nerves. Together, these parasympathetic and sympathetic nerves form the autonomic nerve plexus of the small pelvis (pelvic plexus or inferior hypogastric plexus). The pelvic plexus is a coarse and flat meshwork that is situated laterally to the pelvic organs and supplies the rectum, uterus, vagina, vestibular bulbs, clitoris, bladder, urethra, penis, and prostate. 8,9 Because of their location, disruption of the pelvic plexus and the pelvic splanchnic nerves may occur frequently during dissection of the lateral planes of the mesorectum deep in the pelvis. 10 Parasympathetic injury (pelvic splanchnic nerves or pelvic plexus) produces a hypo- or acontractile bladder with decreased sensation, causing difficulty in bladder emptying. 5,8 The prevalence of faecal incontinence after low anterior resection with preoperative radiotherapy (PRT) is reported to be as high as 60 percent, and even without PRT, to be as high as 40 percent Damage to the pudendal nerve has been suggested as a cause of faecal incontinence after rectal cancer treatment. 14 Common knowledge among clinicians is that the pudendal nerve innervates the levator ani muscle, which is a striated muscular diaphragm that closes the pelvic cavity. The levator ani muscle is the main pelvic floor muscle and is a crucial component of the urinary and faecal continence system Recent studies have re-emphasised the existence of a separate nerve to the levator ani (the levator ani nerve [LAN]), which arises from sacral nerves S3 and/or S4, separately from the pudendal nerve. The nerve is mentioned in various anatomy textbooks 18,19 but is still not clearly illustrated in others. 20,21 The LAN approaches the levator ani muscle from within the pelvis on the superior surface of the pelvic floor, which makes ac- 121

4 Chapter 8 cidental disruption of the nerve during pelvic surgical interventions conceivable This is in contrast with the pudendal nerve, which runs inferior to the pelvic floor muscles and has only a minor contribution to the levator ani muscle innervation. 25 We hypothesised that surgical disruption of the LAN during TME could play a role in the etiology of faecal and urinary incontinence after TME. We aimed to study possible nerve disruption during the TME procedure as a cause of postoperative anorectal and urinary dysfunction. To do this, we performed TME on cadaver pelves and studied the anatomy of the levator ani muscle innervation and its relation to the pelvic autonomic innervation and the mesorectum. Subsequently, data from the Dutch TME trial were analysed to relate anorectal and urinary dysfunction to possible peroperative nerve damage. 26 METHODS Anatomy and cadaver surgery Ten pelves of male cadavers (age range: 67 to 91 years) without signs of pelvic surgery were dissected as described elsewhere 24 to investigate and quantify the topographical anatomy of the LANs and the pelvic splanchnic nerves and the relation to the mesorectal fascia. Total mesorectal excision was performed on one midsagittally transected right male hemipelvis, two complete male pelves, and one complete female pelvis by an experienced colorectal surgeon (B.A.B.). The procedure was performed as it would be in a patient with a low rectal tumour with an indication for a low anterior resection. The rectum was removed according to the TME principles. 27 The pelvic splanchnic nerves and the LAN were subsequently dissected with special reference to relations between the nerves and the rectum, the parietal pelvic fascia, and the mesorectal fascia. Dutch TME trial database Data were obtained from the database of the Dutch TME trial. 26 Between January 1996 and December 1999, patients with histologically proven adenocarcinoma of the rectum and without evidence of distant metastases were included in the trial and randomly assigned to undergo TME alone or preceded by PRT. All patients underwent surgery according to the TME principles, as advocated by Heald. 27 Only patients who underwent low anterior resection and participated in the quality-oflife study or in the study on long-term functional outcome were selected for this analysis. 13,28 122

5 The role of levator ani nerve damage in the development of incontinence problems Outcome measures Faecal and urinary incontinence and difficulty in bladder emptying were evaluated preoperatively and at 5 years after TME. 13,28 To determine whether preoperative dysfunction should be taken into account, the influence of preoperative dysfunction on faecal and urinary incontinence and difficulty in bladder emptying was evaluated. Newly developed combined faecal and urinary incontinence was assumed to result from pelvic floor dysfunction, as this was the most probable causative factor of both dysfunctions. Therefore, to evaluate to what extent incontinence could be explained by dysfunction of the pelvic floor, faecal and urinary incontinence were related to each other. Subsequently, to estimate the risk of simultaneous damage to the LAN in case of damage to the pelvic splanchnic nerves, faecal and urinary incontinence were related to difficulty in bladder emptying. Statistical Analysis Data were analysed with SPSS version 14.0 for Windows (SPSS, Chicago, IL). The influence of the predictor variables on the risk of the different types of dysfunction after TME was calculated using univariate logistic regression analysis. To examine any independent influence, all variables associated with the specific type of dysfunction (p=0.100 in the univariate regression analysis) were entered into a multivariable logistic regression analysis. P=0.05 was considered statistically significant. RESULTS Anatomy and cadaver surgery On macroscopical dissection, we found the LAN to originate from sacral nerves S3 and/or S4. The nerve was macroscopically detectable in all pelves. During its course on the surface of the coccygeus and levator ani muscle, the LAN runs 4 cm (95% CI, 4 to 5.5 cm) lateral to the midsagittal plane at the level of the ischial spine, 4.5 cm (95% CI, 4 to 5.5 cm) lateral to the tip of the coccyx, and 9 mm (95% CI, 0 to 14 mm) caudal to the ischial spine. In all cases, the nerve was situated underneath the pelvic parietal fascia, which covers the levator ani muscle. When the mesorectum was dissected in the surgical plane between the mesorectal fascia and the pelvic parietal fascia and subsequently lifted, the origin of the pelvic splanchnic nerves and the LAN presented themselves as joint sacral branches. The pelvic splanchnic nerves, whose origin from the sacral nerve plexus lies underneath the pelvic parietal fascia, 123

6 Chapter 8 Figure 1. Total mesorectal excision (TME) on an adult male pelvis. The pelvis was midsagittally transected after TME. Note that the parietal fascia was removed from the surface of the levator ani muscle during the TME procedure (black arrows show the cut border). The part of the parietal fascia that covered the sacral origin of the levator ani nerve and pelvic splanchnic nerves was removed during dissection to visualise the nerves. The most distal part of the levator ani nerve was disrupted during TME in this pelvis (not visible). (A) Medial view of the right hemi pelvis with part of the parietal fascia still covering the levator ani nerve. (B) Medial view of the right hemi pelvis. The parietal fascia, covering the nerves, is now flipped sideways to fully reveal the levator ani nerve. A, anus; BL, bladder; P, prostate; Pe, peritoneum; PF, parietal fascia; PP, pelvic plexus; PS, pubic symphysis; R, rectum; S, sacrum; S2, S3, S4, sacral nerves S1 to S3. Figure 2. The levator ani nerve in vivo during total mesorectal excision. (A) Overview. (B) Detail of the levator ani nerve (white arrowheads). BL, urinary bladder; PS, pubic symphysis; SP, sacral promontory. run in a separate fascial sheath to reach the pelvic plexus that is situated lateral to the rectum, tangentially to the lateral surface of the mesorectal fascia. The LAN does not run in this fascial sheath, but continues solitary toward the pelvic floor muscles underneath the lateral pelvic parietal fascia. Dissection of the LAN in the pelves after TME showed a similar anatomic composition. The plane of surgical dissection during TME is between the visceral fascia of the mesorectum and the parietal fascia. At the most distal part of the rectum, approximately 2 cm cranial from the entrance 124

7 The role of levator ani nerve damage in the development of incontinence problems Figure 3. Anatomic relation of the mesorectum and the levator ani nerve in the adult male. (A) Transverse section through the male pelvis. The level of the section is illustrated in the inset. (B) Schematic overview of the structures shown in part A. The green circles represent the approximate position of the levator ani nerves. (C, D) Three-dimensional reconstruction of the same adult pelvis as in part A with (D) and without (C) the mesorectum. The levator ani muscle (red), coccygeus muscle (pink), obturator internus muscle (mint green), rectum (blue), mesorectum (light blue), sacral nerve plexus (light green) are reconstructed from serial sections. Because the levator ani nerve (light green, arrowheads in C) was not identifiable in the serial sections, it was illustrated with information from our dissection studies. Note the close relation of the mesorectum to the levator ani nerves. BL, urinary bladder; BP, bony pelvis; C, coccyx; CM, coccygeus muscle; GM, gluteus maximus muscle; LA, levator ani muscle; MR, mesorectum; OI, obturator internus muscle; R, rectum; SV, seminal vesicle. of the rectum through the levator ani muscle, the mesorectal fascia and the parietal fascia become inseparable. Therefore, the parietal fascia must be removed from the surface of the pelvic floor muscles to preserve the mesorectal package. At this point, the LAN is in close proximity of the surgical dissection plane. In one male pelvis, the LAN was disrupted unilaterally at this level during the TME procedure. In the 125

8 Chapter 8 pelvis where the LAN was disrupted, Figure 1 shows the close relation between the surgical dissection plane and the nerve in the lowest part of the dissection. Figure 2 shows the LAN in vivo during TME. Figure 3 illustrates the close relation between the LAN and the mesorectum. Dutch TME trial database Of the Dutch patients, patients were excluded from analysis for the following reasons: ineligible at randomisation (n=50), no operation (n=37), in-hospital death (n=52), no informed consent for the quality-of-life study (n=89), no quality-of-life forms returned (n=30), no low anterior resection (n=456), and missing pretreatment form (n=165). Consequently, 651 patients remained assessable for analysis. However, not all questions were answered by every patient in the returned questionnaires, resulting in 649, 647, and 649 assessable patients concerning faecal incontinence, urinary incontinence, and difficulty in bladder emptying, respectively. Pre- and postoperative dysfunctions Faecal incontinence was reported by 269 (41.4 percent) of 649 patients preoperatively and by 134 (48.7 percent) of 275 patients 5 years after rectal cancer treatment (p=0.001; relative risk=3.16). Of patients with normal preoperative function, 38.8 percent (69 of 178 patients) had newly developed faecal incontinence after rectal cancer treatment. Urinary incontinence was reported by 110 (17.0 percent) of 647 patients preoperatively and by 123 (39.5 percent) of 311 patients 5 years after rectal cancer treatment (p<0.001; relative risk= 4.19). Of patients with normal preoperative function, 33.7 percent (88 of 261 patients) had newly developed urinary incontinence after rectal cancer treatment. Of patients with newly developed faecal incontinence, 36.5 percent (23 of 63 patients) also reported newly developed urinary incontinence (p=0.027; relative risk=2.21). Fourteen percent (23 of 160) of patients with normal preoperative function reported both faecal and urinary incontinence after rectal cancer treatment. Difficulty in bladder emptying was reported by 144 (22.2 percent) of 649 patients preoperatively and by 89 (29.3 percent) of 304 patients 5 years after rectal cancer treatment (p=0.001; relative risk=3.02). Of patients with normal preoperative function, 24.3 percent (59 of 243 patients) had newly developed difficulty in bladder emptying after rectal cancer treatment. Of patients with newly developed faecal and urinary incontinence, 38.8 percent (7 of 18 patients) also reported difficulty in bladder emptying (p=0.044; relative risk=2.34). 126

9 The role of levator ani nerve damage in the development of incontinence problems Risk factors Table 1 lists the results of the univariate logistic regression analysis of risk factors for faecal and urinary incontinence and difficulty in bladder emptying. In the multivariable analysis of faecal incontinence (Table 2), PRT, low anastomotic height, tumour Table 1. Logistic regression analyses of risk factors for faecal and urinary incontinence and difficulty in bladder emptying after rectal cancer treatment (RR=relative risk) Proportion with dysfunction Faecal incontinence Urinary incontinence Difficulty in bladder emptying RR p-value Proportion with dysfunction RR p-value Proportion with dysfunction RR p-value Gender male female 80 of of of of < of of Age < of of of of of of Preoperative dysfunction no yes 69 of of < of of < of of <0.001 Radiotherapy no yes 56 of of of of of of Posterior tumour location no yes 54 of of of of of of Tumour size (cm) < of of of of of of Blood loss (ml) < of of of of of of Anastomotic height (cm) < of of of of of of Anastomotic leakage no yes 126 of of of of of of

10 Chapter 8 Table 2. Multivariable regression analyses of risk factors for faecal and urinary incontinence and difficulty in bladder emptying after rectal cancer treatment (RR=relative risk) Faecal incontinence Urinary incontinence Difficulty in bladder emptying RR p-value RR p-value RR p-value Preoperative dysfunction 3.48 < <0.001 Female gender Age Tumour size 4.0cm Radiotherapy Blood loss 1500ml Anastomotic height <6.0cm size (p<0.100 in the univariate analysis), and preoperative faecal incontinence were included as input variables to predict faecal incontinence after rectal cancer treatment. In the multivariable analysis, PRT (p=0.004; relative risk=2.25) and preoperative faecal incontinence (p=0.001; relative risk=3.48) remained significant predictors. In the multivariable analysis of urinary incontinence (Table 2), female gender (p=0.001; relative risk=2.34), age (p=0.014; relative risk=1.96), preoperative urinary incontinence (p=0.017; relative risk=2.44), and low anastomotic height (p=0.049; relative risk=1.72) remained significant predictors. In the multivariable analysis of difficulty in bladder emptying (Table 2), preoperative dysfunction (p<0.001; relative risk=3.15) and excessive peroperative blood loss (p=0.038; relative risk=1.95) remained significant risk factors. DISCUSSION The present study aimed to evaluate nerve disruption during TME as a cause of poor functional outcome by using anatomic and clinical data, with special attention to LAN and incontinence. Fourteen percent of patients newly developed combined faecal and urinary incontinence after TME and, therefore, probably had a dysfunctional pelvic floor. As the cadaver surgery study revealed, the nerve supply to the pelvic floor, by means of the LAN, lies in the field of operation and can be disrupted during TME. From the anatomic findings, it can be predicted that especially during TME for low tumours, where the parietal fascia of the levator ani muscle is entered, the LAN is at risk. Indeed, our TME database analysis demonstrates that an anastomotic level less than 6 cm increased the risk of (combined) faecal and urinary incontinence significantly. 128

11 The role of levator ani nerve damage in the development of incontinence problems Additionally, in other studies, low anastomotic level, next to PRT, is considered to be the most important risk factor for faecal incontinence. 11,29,30 We found that the anatomic origin of the LAN was closely related to the origin of the pelvic splanchnic nerves. From this, it can be predicted that improper surgical dissection or excessive manipulation of the mesorectum hold a risk of combined disruption of the LAN and the pelvic splanchnic nerves. In one third of patients, newly developed faecal and urinary incontinence was accompanied by newly developed difficulty in bladder emptying. This would imply that in one third of patients, in whom the LAN was disrupted, simultaneous disruption of the LAN and pelvic splanchnic nerves occurred at the sacral origin. The TME dissection plane along the parietal presacral fascia is likely to mislead the surgeon and result in injury to the pelvic splanchnic nerves and/or pelvic plexus because the parietal presacral fascia divides into several laminae lining or enclosing these nerves. In addition, when an incorrect plane is followed, the sacral venous plexus, which lies in close proximity of the pelvic splanchnic nerves, may be damaged, resulting in excessive blood loss. 31 Indeed, excessive peroperative blood loss was significantly associated with difficulty in bladder emptying in our study. Diathermic coagulation and numerous sutures to secure hemostasis may cause nerve damage. Moreover, excessive blood loss hinders sight deep in the pelvis, making nerve sparing virtually impossible. 3,32 An increased risk of nerve damage and poor functional outcome in case of a posteriorly located tumour would be expected. However, this is not supported in the present study. Apparently, surgical damage during TME does not depend on characteristics of the tumour but only on specific aspects of the surgical technique used. Faecal incontinence is multifactorial. 33 The rectum acts as a reservoir, and the smaller neorectum after TME has a lower capacity and smaller tolerated volume. 34 Furthermore, PRT is known to increase the risk of faecal incontinence, which is also supported by the present study. 11,35 Radiotherapy diminishes compliance of the residual rectum because of fibrosis and may disrupt the myenteric plexus of the internal anal sphincter, compromising the rectoinhibitory reflex and resting anal pressures. 33,34 In addition, faecal incontinence after rectal cancer treatment has been reported to be caused mainly by impaired pelvic floor movement (ie, a disturbed change in anorectal angle resulting from a dysfunctional puborectalis muscle). 36 Urinary incontinence after TME is multifactorial as well. Unfortunately, the questionnaires did not differentiate between urge, overflow, and stress incontinence. Damage to the sympathetic nerve supply (the hypogastric nerves and the pelvic plexus) causes a reduced bladder capacity and may result in urge incontinence. 5 However, one-sided preservation of the pelvic plexus has been clinically shown to result in acceptable urinary continence. 3,37 Surgeons are often unable to verify whether 129

12 Chapter 8 bilateral damage has occurred, but it is not expected to occur frequently during a TME procedure. 4 Damage to the sacral splanchnic nerves may lead to difficulty in bladder emptying and overflow incontinence. 5 However, urinary incontinence was not significantly related to difficulty in bladder emptying in this study (data not shown). Therefore, we assume that the reported urinary incontinence was mainly stress incontinence. Stress incontinence may also result from impaired support to the urethra and bladder neck. 38 As for faecal incontinence, a dysfunctional pelvic floor has been suggested as an important cause of urinary incontinence This is supported by our results, as faecal and urinary incontinence were occurring simultaneously in a significant number of patients. In conclusion, the results of our study lead us to state that, especially in patients with low rectal tumours, the risk of LAN disruption is substantial, which could contribute to an increased risk of urinary and faecal incontinence after TME, as indicated by our clinical data. Accidental disruption of the LAN during a surgically difficult procedure could be a factor that has been neglected thus far. The results of our surgical study imply that a correctly performed posterior dissection of the mesorectum would not disrupt the LAN, because the plane of posterior dissection in a TME procedure is between the pelvic parietal fascia and the mesorectal fascia. However, the surgical margin is so small that any deviation from this surgical plane easily results in disruption of the nerve. Adhering to the surgical plane, reducing the use of blunt dissection, and improving rectal retraction may lower the risk of LAN disruption during distal resection. Surgeons that perform TME should be aware of the anatomy of the LAN to avoid disrupting it. A nerve sparing TME should mean not only sparing the pelvic autonomic nerves, but sparing the LAN as well. The challenge is now to assess puborectalis function in patients suffering from faecal incontinence after TME, to actually see whether the puborectalis muscle is denervated. Further studies on faecal incontinence after TME should therefore include clinical assessment of pelvic floor denervation (ie, puborectalis muscle atrophy) in patients who suffer from faecal incontinence after TME. 130

13 The role of levator ani nerve damage in the development of incontinence problems REFERENCE LIST 1. Havenga K, Enker WE: Autonomic nerve preserving total mesorectal excision. Surg Clin North Am 82: , MacFarlane JK, Ryall RD, Heald RJ: Mesorectal excision for rectal cancer. Lancet 341: , Moriya Y: Function preservation in rectal cancer surgery. Int J Clin Oncol 11: , Maas CP, Moriya Y, Steup WH et al: A prospective study on radical and nerve preserving surgery for rectal cancer in the Netherlands. Eur J Surg Oncol 26: , Havenga K, Maas CP, DeRuiter MC et al: Avoiding long-term disturbance to bladder and sexual function in pelvic surgery, particularly with rectal cancer. Semin Surg Oncol 18: , Maas CP, Moriya Y, Steup WH et al: Radical and nerve-preserving surgery for rectal cancer in The Netherlands: A prospective study on morbidity and functional outcome. Br J Surg 85:92-97, Nesbakken A, Nygaard K, Bull-Njaa T et al: Bladder and sexual dysfunction after mesorectal excision for rectal cancer. Br J Surg 87: , Maas CP, De Ruiter MC, Kenter GG et al: The inferior hypogastric plexus in gynecologic surgery. J Gynecol Technique 5:55-62, Ali M, Johnson IP, Hobson J et al: Anatomy of the pelvic plexus and innervation of the prostate gland. Clin Anat 17: , Rees PM, Fowler CJ, Maas CP: Sexual function in men and women with neurological disorders. Lancet 369: , Bretagnol F, Troubat H, Laurent C et al: Long-term functional results after sphincter-saving resection for rectal cancer. Gastroenterol Clin Biol 28: , Lange MM, den Dulk M, Bossema ER et al: Risk factors for faecal incontinence after rectal cancer treatment. Br J Surg 94: , Peeters KC, van de Velde CJ, Leer JW et al: Late side effects of short-course preoperative radiotherapy combined with total mesorectal excision for rectal cancer: Increased bowel dysfunction in irradiated patients A Dutch colorectal cancer group study. J Clin Oncol 23: , Tomita R, Igarashi S, Ikeda T et al: Pudendal nerve terminal motor latency in patients with or without soiling 5 years or more after low anterior resection for lower rectal cancer. World J Surg 31: , Ashton-Miller JA, DeLancey JO: Functional anatomy of the female pelvic floor. Ann N Y Acad Sci 1101: , Madoff RD, Parker SC, Varma MG et al: Faecal incontinence in adults. Lancet 364: , Norton P, Brubaker L: Urinary incontinence in women. Lancet 367:57-67, Drake RL, Vogl W, Mitchell AWM: Gray s Anatomy for Students. Philadelphia, PA, Elsevier, Netter FH: Atlas of Human Anatomy (ed 2). Teterboro, NJ, ICON Learning Systems, Martini FH, Timmons MJ, Tallitsch RB: Human Anatomy (ed 5). Upper Saddle River, NJ, Prentice Hall Pearson Education,

14 Chapter Putz R, Pabst R: Sobotta Atlas of Human Anatomy (ed 13). Philadelphia, PA, Lippincott Williams & Wilkins, Azpiroz F, Fernandez-Fraga X, Merletti R et al: The puborectalis muscle. Neurogastroenterol Motil 17:68-72, 2005 (suppl 1) 23. Barber MD, Bremer RE, Thor KB et al: Innervation of the female levator ani muscles. Am J Obstet Gynecol 187:64-71, Wallner C, Maas CP, Dabhoiwala NF et al: Innervation of the pelvic floor muscles: A reappraisal for the levator ani nerve. Obstet Gynecol 108: , Wallner C, van Wissen J, Maas CP et al: The contribution of the Levator Ani Nerve and the Pudendal Nerve to the innervation of the Levator Ani Muscles: A study in human fetuses. Eur Urol /j.eururo [epub ahead of print on November 20, 2007] 26. Kapiteijn E, Marijnen CA, Nagtegaal ID et al: Preoperative radiotherapy combined with total mesorectal excision for resectable rectal cancer. N Engl J Med 345: , Heald RJ: Rectal cancer: The surgical options. Eur J Cancer 31A: , Marijnen CA, van de Velde CJ, Putter H et al: Impact of short-term preoperative radiotherapy on health-related quality of life and sexual functioning in primary rectal cancer: Report of a multicenter randomized trial. J Clin Oncol 23: , Guren MG, Eriksen MT, Wiig JN et al: Quality of life and functional outcome following anterior or abdominoperineal resection for rectal cancer. Eur J Surg Oncol 31: , Welsh FK, McFall M, Mitchell G et al: Preoperative short-course radiotherapy is associated with faecal incontinence after anterior resection. Colorectal Dis 5: , Baqué P, Karimdjee B, Iannelli A et al: Anatomy of the presacral venous plexus: Implications for rectal surgery. Surg Radiol Anat 26: , Junginger T, Kneist W, Heintz A: Influence of identification and preservation of pelvic autonomic nerves in rectal cancer surgery on bladder dysfunction after total mesorectal excision. Dis Colon Rectum 46: , Putta S, Andreyev HJ: Faecal incontinence: A late side-effect of pelvic radiotherapy. Clin Oncol (R Coll Radiol) 17: , Varma JS, Smith AN, Busuttil A: Correlation of clinical and manometric abnormalities of rectal function following chronic radiation injury. Br J Surg 72: , Pollack J, Holm T, Cedermark B et al: Longterm effect of preoperative radiation therapy on anorectal function. Dis Colon Rectum 49: , Köninger JS, Butters M, Redecke JD et al: Transverse coloplasty pouch after total mesorectal excision: Functional assessment of evacuation. Dis Colon Rectum 47: , Matsuoka N, Moriya Y, Akasu T et al: Longterm outcome of urinary function after extended lymphadenectomy in patients with distal rectal cancer. Eur J Surg Oncol 27: , Ulmsten U: Some reflections and hypotheses on the pathophysiology of female urinary incontinence. Acta Obstet Gynecol Scand Suppl 166:3-8,